Effects of insect herbivores on individual plant performance have been demonstrated in many systems. In contrast, the possibility that these individual-level effects alter plant population dynamics has not been well tested. Such tests are important because the hypothesis that herbivores affect population dynamics underpins fundamental and applied issues including biological control and the enemy release hypothesis. A complete test of herbivore effects on plant population size requires accounting for density dependence in the plant population. Here, we ask how insect herbivores affect density-dependence and population dynamics in the early-successional perennial weed Solanum carolinense. We use four years of data from a long-term field experiment manipulating both plant density (five initial levels) and herbivores (present vs. absent). We monitored herbivore damage levels as well as plant survival, growth, and reproduction in 40 experimental plots (n=5 per density-herbivore treatment combination). We tested for density dependence in damage levels and herbivores, and examined effects on plant size. We also used demographic data to parameterize stage-structured matrix population models for S. carolinense. We estimated transitions between stages and calculated population growth rate (lambda) in the presence and absence of herbivores, and for plots at different densities. We then tested for density dependence in transitions between size classes as well as differences in density dependence for plots with and without herbivores.
Results/Conclusions
Although herbivore observations did not vary among plots at different densities, we did detect some density effects on damage levels. Herbivory and S. carolinense density both affected stem size in some years. They also affected population growth rate, but effects varied with year. Early on, population growth rate was negatively affected by density in the absence but not presence of herbivores. In subsequent years, either density or herbivory alone affected population growth rate, but density effects were not altered by herbivory. In each year, density affected some transitions; which transitions were affected varied both by year and by herbivore treatment. Together, these results suggest that herbivores may affect plant population dynamics, but the effects likely vary through time. In particular, herbivore effects may change during succession. For early successional species like S. carolinense, herbivores likely have greater effects early in succession, when damage has stronger effects on density dependence.